Many of us have seen films containing remarkably realistic dinosaurs, aliens, animated toys and other fanciful creatures. Such animations are made possible by computer graphics. Using such techniques, a computer graphics artist can specify how each object should look and how it should change in appearance over time, and a computer then models the objects and displays them on a display such as your television or a computer screen. The computer takes care of performing the many tasks required to make sure that each part of the displayed image is colored and shaped just right based on the position and orientation of each object in a scene, the direction in which light seems to strike each object, the surface texture of each object, and other factors.
Because computer graphics generation is complex, computer-generated three-dimensional graphics just a few years ago were mostly limited to expensive specialized flight simulators, high-end graphics workstations and supercomputers. The public saw some of the images generated by these computer systems in movies and expensive television advertisements, but most of us couldn't actually interact with the computers doing the graphics generation. All this has changed with the availability of relatively inexpensive 3D graphics platforms such as, for example, the NINTENDO 64®, the NINTENDO GAMECUBE® and various 3D graphics cards now available for personal computers. It is now possible to interact with exciting 3D animations and simulations on relatively inexpensive computer graphics systems in your home or office.
A problem graphics system designers have confronted is how to efficiently model and render realistic looking animations in real time or close to real time. To achieve more interesting dynamic animation, a number of video and computer games have used various animation techniques such as key frame transformations, inverse kinematics and the like to model and animate people, animals and other objects. See for example O'Rourke, Principles of Three-Dimensional Computer Animation (W. W. Norton 1998) at chapters 3 and 4 especially. While such techniques have been highly successful, animators have searched for ways to make animations more realistic without the need to control or map out each and every movement of an animated character beforehand.
One approach that has appeal is to make the animation engine responsible for making animated characters increasingly more intelligent. For example, it is possible to define an “intelligent” animated character within a three-dimensional environment and allow the character to react to the environment based on its programmed qualities. If the character is sufficiently intelligent, rather complex reactions can be dynamically created “on the fly” by the real time animation engine—saving the game developer the massive amount of time and effort that might otherwise be required to script out the animation sequence. See, for example, U.S. patent application Ser. No. 09/382,819 of Comair et al filed Aug. 25, 1999 entitled “Object Modeling For Computer Simulation And Animation” incorporated by reference herein.
While such approaches have been successful, further improvements are possible. In particular, we have developed a new, efficient technique for causing an animated character to pay attention to an object within a virtual world by tagging the object. When the animated character moves into proximity with an object (e.g., in response to user control), the system checks whether the object is tagged. If the object is tagged, the animation engine animates the character to pay attention to the tagged object (e.g., by animating the character to look or stare at the tagged object so long as the character remains close to the tagged object). The tag-based animation engine can, for example, animate the character to turn toward or face the tagged object in the process of paying attention to it—creating a very realistic visual effect without the typical programming overhead normally required to specify which direction the animated character should face and when. In other words, in accordance with the invention, the tags can be defined by designers at any location in the virtual world and given certain characteristics that are designed to cause a character that comes into a defined proximity of the tag to have some sort of reaction to the tag. By defining several tags in a scene, such as, for example, in a virtual hallway through which a character is walking, the animation engine makes the character much more realistic and to appear as if the character is coming to life through its reactions to the tags. The tags are preferably associated with virtual objects of the type that would typically cause a human to have a reaction in the real world. The tags are preferably defined to cause the same type of reaction in the character's animation, as a typical human would have in the same circumstance in the real world. In this way, the character has much more human-like reactions to its environment while moving through the virtual world, and the character can be made to appear as if it has “come to life.”
In accordance with the invention, the tagging system of the animation engine is preferably priority based. In other words, each tag is assigned a priority value that is used by the animation engine to control which tag will be used when more than one tag is active. By prioritizing the tags in the environment, the animation engine as able to display the character as paying attention to or reacting to the tagged object that is of highest interest to the character, based on the character's current environment and/or state, from among several tags that may be in proximity to the character at any one time. This tag prioritization feature further helps to make the character appear more realistic by enabling the character to prioritize its reactions in the same or similar way to that of a human. For example, in the real world, humans typically are confronted with numerous objects (e.g., interesting painting, view, other object etc.) or events (loud noise, flashing light, movement, etc.) that may cause a reaction at any one time. However, humans, by their nature, typically react to the one thing that seems to be the most important at each instant in time. For instance, a human would typically stop looking at a piece of art when a loud noise comes from another object, and then quickly turn in the direction of the loud noise. Upon determining that the noise is not a problem, a human would then typically resume looking at the piece of art. These same human-like movements and reactions can be generated by the reactive animation system of the invention, by giving the object that makes the noise a higher priority tag while active as compared to the tag associated with the piece of art. In this way, all of the tagged objects in the environment can have relative priorities assigned thereto based on, for example, the nature of the object.
In one particular embodiment, the object can be tagged with a tag that inspires an emotion in the character while paying attention to the tagged object. The emotion can, for example, be fear, happiness, or any other discernible emotion. If an object is tagged to inspire fear, the character can be animated to turn toward the object and react with a look of horror. If an object is tagged to inspire happiness, the character can be animated to turn toward the object and react with a big smile. Other emotions and reactions are possible. In fact, the tag can be defined to cause any type of response that corresponds to any variable or role-playing element that the character may have, as well as to cause emotional and/or physical reactions. For example, the tag could modify the animation of the character so that the character appears injured, sick or insane while under the influence of an active tag.
In accordance with the invention, the character's animation is adapted to the tag when the tag is activated. Activation of the tag can occur when the user gets within a selected distance from the tag and/or based on some other defined event. The adaptation of the animation is preferably done by defining key frames for use in creating a dynamic animation sequence using the information provided by the tag. The dynamic animation sequence is preferably generated using the techniques known in the art as “Inbetweening” and “Inverse Kinematics.” Inbetweening enables the frames between the key frames to be generated for the dynamic animation, and inverse kinematics is used to assure that the character's movements are natural during the animation. Once a tag is activated, the animation of the character is adapted from its generic or canned-animation to a dynamic animation based on the type of tag that has been encountered. Thus, the tag triggers a dynamic modification of the character's animation for the period of time it takes for the tag to become inactive, such as by the character moving out of range of the tag.
The dynamic animation provided by the reactive animation engine of the instant invention provides a character (or other object) with realistic reactions as the character moves through the virtual environment, without having to handcraft the animation for every possible scene in advance. Thus, the invention enables animation to be generated on-the-fly and in an unpredictable and realistic manner. As a result, the character's animation can constantly change in a variety of ways and depending on many possible variables. This makes the character's animation unpredictable and greatly enhances the visual effect of the display. A significant advantage of the instant invention is that the character can be displayed with a myriad of animations without having to script, hand-craft or store each of the animations in advance. Instead, the reactive animations are dynamically generated on-the-fly and in real time based on the tag.